Skip to content

When you choose to publish with PLOS, your research makes an impact. Make your work accessible to all, without restrictions, and accelerate scientific discovery with options like preprints and published peer review that make your work more Open.


Unleashing the Power of Collaboration: Open Science on Earth Day

Today, April 22nd, we celebrate Earth Day, a critical reminder of the need to protect and preserve our planet. It serves as a rallying call for individuals and organizations to take action towards addressing environmental challenges, such as the climate emergency, biodiversity loss, and pollution.

Open science, which promotes transparency, collaboration, and accessibility in scientific research, can play a crucial role in supporting the goals of Earth Day. By making scientific knowledge and data freely available to the public, open science can facilitate informed decision-making, foster interdisciplinary collaborations, and promote evidence-based policies for environmental conservation and sustainability. It can also enable citizen science initiatives, empowering individuals and communities to actively engage in environmental monitoring, research, and advocacy.

Two recent publications in PLOS ONE elegantly illustrate how open science practices can unlock new research opportunities and deliver outputs that are as useful as possible for practitioners and policymakers. 

In the first article, published in March by researchers from the USA, Chile, Australia, and Sweden, the study team examined a huge dataset on plastic pollution in the world’s oceans. Collected from more than 11,000 stations over 40 years, the data allowed the authors to explore trends in global abundance of ocean plastics, with the staggering conclusion that there are now likely over 170 trillion plastic particles now floating in the world’s oceans. Even more striking is the pattern over time, with little observable trend in abundance before 2005 but a rapid increase thereafter. To assemble the full dataset, the authors had to collate data from 17 published sources (including academic articles and public repositories) – an endeavor that would not have been possible without those previous publications making their data available to future researchers. And, to their credit, the authors have followed in their predecessors’ footsteps, publishing the complete dataset and accompanying code on Github

In the second article, published earlier this month by researchers based in Mauritius, the authors report the design and validation of a deep learning model to identify crown-of-thorns starfish, a major coral predator, in images of coral reefs. Existing techniques for detecting these starfish used so-called ‘black-box’ models, which output a prediction with no information on the factors which the system used to make its decision. This lack of transparency can make it more challenging to assess and mitigate potential biases, errors, or unintended consequences in the context of complex environmental systems. In the PLOS ONE study, however, the researchers designed a system which outputs both a prediction of whether an image contains a crown-of-thorns starfish and a list of the most important features of the image which the model used to make its decision. Combined with a high classification accuracy of 93%, this increased transparency means the model would be more likely to be used in real environmental management settings.

With each passing year, the need for collective action and global cooperation to safeguard our environment has become increasingly imperative, making Earth Day’s messages all the more relevant and vital in the present times. Open science makes new research possible, maximizes the value of research to tackle environmental challenges, and makes science truly accessible to all: contributing to the collective effort of safeguarding our planet for current and future generations.

Featured image: photo by Noah Buscher on Unsplash

Leave a Reply

Your email address will not be published. Required fields are marked *

Add your ORCID here. (e.g. 0000-0002-7299-680X)

Related Posts
Back to top